In our increasingly environmentally conscious world, making data centres more energy-efficient and sustainable is a huge challenge.
Large server factories used to store, manage and recover our data are extremely power hungry. Not only do they require a huge amount of electricity to operate, they also use up energy for cooling systems used to minimize overheating. “Many data centres already consume as much energy as small cities,” agrees Jay Taylor who heads a subcommittee (SC) inside the joint technical committee (JTC 1) set up between IEC and ISO to establish standards relating to IT.
SC 39 was set up to prepare standards on sustainability for and by IT. It initially worked on a global power effectivenes measurement tool called power usage effectiveness (PUE). The work on PUE led to the publication of ISO/IEC 19395 which facilitates the resource monitoring and control of smart data centres. “I consider it my job to provide practical and usable tools to be able to assess large energy consumers such as data centres. These tools can then be used to reduce our impact on the environment,” Taylor says.
Other publications issued by SC 39 include the ISO/IEC 22237 series of technical specifications. “I would describe them as a set of guidelines which provide data centre owners and operators with a practical methodology they can use when they set up in different countries across the globe, for instance in earthquake-prone zones,” Taylor adds.
The SC is currently working on new energy efficiency specifications for server storage systems. Zombie servers which suck up power without doing any useful work can be an aspect of the less efficient and more ancient data centres. Theses energy wasters are more difficult to find in the more recent server factories that use a uniform computing architecture and which can scale up to thousands of servers. But nevertheless energy efficiency adjustments can be made.
,”The idea is to look at which servers are idle and to switch them off at certain periods of the day or night in order to save energy. Algorithms can be used to predict peak usage, when maximum server power is required. They can shut servers down remotely in off-peak periods. One of the problems we have, however, is that unlike your average PC or laptop, servers are never totally idle even when they are not actively processing information.” CBI